HN user

pragmar

234 karma

Software developer, located in New Hampshire.

Currently working on http://interro.bot

Posts10
Comments103
View on HN

I'm distributing msix (msixbundle) because it allows me to package a downloadable installer and a windows store upload. You should be able to double click the msixbundle to run the installer, powershell isn't necessary.

MSIX installers feel a bit odd, I think, because they're far less common than either msi or self extracting exe. They also virtualize the app. MSIX apps, for example, can't write to the canonical registry. It's not all upside, but it works well for my projects.

The SDK install is expected (if not already installed). It should hopefully have been automatically downloaded during the install process.

I don't mean to imply it's not possible, but I ran into issues packaging with Desktop Bridge and the web started pushing me toward extreme solutions, so I changed course. If there's a lesson here, it's to package early (and know what you are up against).

Apple locks users in with software/services. nVidia locks in add-in board manufacturers with exclusive arrangements and partner programs that tie access to chips to contracts that prioritize nVidia. It happens upstream of the consumer. It's always a matter of degree with this stuff as to where it becomes anti-trust, but in this case it's overt enough for governments to take notice.

Became unusable for me (Firefox/Win) a few months back on my laptop with integrated intel graphics. Fine on other computers with more capable hardware. The solution was an extension called h264ify. There was something about the vp8/9 codec that was bringing Firefox to its knees. h264ify avoids those codecs altogether on Youtube.

This project started back in the day as native MVVM (UWP). Switched to Blazor Server/electron a few years back. For Maui, I had to move to Hybrid but I'm glad I did. Hybrid is pretty great, seems better in every way except routing and some aspnet functionality is lost. Faster app startups, and feels generally more responsive.

Avalonia looks pretty nice, but I've never tried it so I'm not sure how it compares.

I've been porting a Blazor/.net 8 project (https://interro.bot) to Maui for the past couple months. I just released Android last week, and Mac/iOS is in the pipeline. There are plenty of gotchas and I wouldn't say Maui feels mature yet, but the issues in my case are mostly on the build side. I'm optimistic though, the platform does improve with regular releases, and the end product (once through the build headaches) is solid.

I'm not sure I've the best antibot solution, but it's handled through some crawler options exposed to the user. Out of the box, I use a (fast) http crawler with my app's user-agent. It is not at all resilient to antibot. I direct users who are encountering issues to first try a user-agent override, and if that doesn't work, to next enable javascript crawling (think headless chrome), which is slower and heavier, but clears up a lot of issues. I don't have a strategy for aggressive antibot (captcha/etc.) other than to tell the customer to dial it back on their website.

Edit: I've seen antibot SAAS providers, which claim to provide workarounds at a cost. You route traffic through their network, and they have teams that are constantly tweaking things to keep the requests working, much like scrapers adapting to website redesigns. It would be a treadmill to do on your own. There more info on this at https://substack.thewebscraping.club/ In my case, selling single-user perpetual licenses, it doesn't make sense.

Congratulations on shipping, the site looks great. I've been working on a desktop app in the same space (website health/administration). It's been a similarly long journey. I've found marketing to be an uphill battle. It's a crowded marketplace, which in many ways indicates a healthy ecosystem, but difficult to crack nonetheless. Having a focused audience (e.g. bloggers) will help there. If you ever want to discuss marketing or talk shop, definitely reach out (ben at hn-username dot com)... maybe we could both learn something.

I remember Nate Silver speaking of the golden age of online poker, when amateurs were the dominant audience, and a decent (non-pro) strategy player could make a living. Word got out, and just like that, it was over.

Best feature of Pro is Windows Sandbox, which I use to test/trial app installs in a VM. It's basically a disposable OS that spins up as fast as most apps. Sandbox runs on Hyper-V, which along with domain networking (for business) are the two main differentiators as far as I understand.

The Upwork job posting asks the applicant to "fix bugs and resopnsiveness [sic] on website" and claims to pay between $15 and $20 hourly for a task expected to take under a month.

A take-home for a web3 dev-job that pays less than California's minimum wage (as of April 2024) is a story in itself.

Just one opinion--but as long as you're offering a free beta, you shouldn't worry about it. Getting users in the loop early is usually the better option. As for coping with rejection/insecurities, it gets easier with time and practice. Just remember that it's nothing personal. We're all bombarded with commercial messaging, and your cold email is just that.

In the aughts, when everyone was rolling their own framework, security and maintenance were a nightmare. It's undeniable. We traded one problem for another, however, and we've gone too far. I think the question at hand is which is more secure/maintainable: 10kb of custom utilities or 100mb of mystery-meat modules that, let's face it, will never be reviewed. It's not a simple answer.

It's not statistically incorrect. KPIs are generally discussed in time frames, year over year, quarter over quarter. There is year over year growth in multiple jobholders, three years running. That constitutes a strong trend, even if regressing to the mean. Also, if you're not filtering your multiple jobs data to prime age, you should use aggregate participation rate (unfiltered), which tells a different story.[0]

[0] https://fred.stlouisfed.org/series/CIVPART

I don't know anyone who can turn it on and off on a dime. But it isn't difficult to turn off in my experience, just takes removing the behavior (side projects) from your routine for about a month. Once off, it's hard to turn on again, which may (or may not) be an acceptable outcome.

Bonds are a good place to hang out in turbulent markets if your goal is to preserve what you have, but be careful with municipal and high-yield ("junk") bonds. These can and do go to zero sometimes. Bond etfs can sting unsuspecting investors, as their value can shift dramatically when rates change. Government treasuries are considered safest.

Civil liberties and democracy are often at odds. Where states that have implemented a ban, it has been done democratically. The fact that it was done in a previous congressional session doesn't change that. Democracy and civil liberties are both great, but in this case it's a "pick one" situation.

The ostensible motivation of a possible US/UK/Polish operation, as I understand it, was poisoning the well on peace negotiations. By removing the optionality of turning the lines back on, there is less for the Germans and Russians to work with, and therefore less to motivate diplomatic talks in the first place. The Germans may be committed to getting off Russian gas permanently, but they still have to get there from here without losing their economy.